ABSTRACT We seek to understand the progression from inflammation to neoplasia in patients with inflammatory bowel disease (IBD). Often patients are diagnosed with dysplasia and undergo colectomies even though they will never develop cancer. The focus of our studies has been the link between intestinal bacteria and innate immune signaling leading to colitis-associated neoplasia. Toll-like receptor 4 (TLR4) recognizes LPS and other endogenous ligands present in the inflamed colon. Our group has shown that TLR4 is overexpressed in inflamed and dysplastic tissues of ulcerative colitis (UC) patients. Using transgenic and knock-out mice, we have shown that intestinal epithelial expression of TLR4 engenders a colitogenic microbiome capable of transmitting inflammation. Mice with constitutively active TLR4 in epithelial cells show dramatically increased epithelial proliferation and increased colonic tumors (AOM-DSS). Our preliminary data demonstrate that cancer stem cells from these tumors have a very distinct transcriptional profile compared to tumors in WT mice. We also show that TLR4-driven tumors have enhanced recruitment of tumor supporting myeloid-derived suppressor cells (MDSCs). TLR4-driven tumors have dramatic upregulation of NADPH oxidases and enhanced production of H2O2. H2O2 and other reactive oxygen species have several biological functions, including control of cell proliferation and microbial populations. In the current proposal, we hypothesize that dysregulation of TLR4 induces redox signaling, promoting stem cell activation and leading to selection of a tumorigenic microbiome. We further hypothesize that the dysbiotic microbiome activates MDSCs to promote CAC. This is pursued in the following specific aims: 1) Analyze the role of TLR4-mediated redox signaling on cancer stem cell activation in CAC. We will use our diverse animal models for TLR4 and the NADPH dual oxidase 2 (Duox2) to define how redox signaling participates in activation and survival of cancer stem cells during TLR4- driven tumorigenesis. 2) Interrogate the inter-relationship of TLR4-mediated redox signaling on mucosal dysbiosis and the consequences of dysbiosis in tumorigenesis. We will use our animal models and metagenomic sequencing to investigate how redox signaling through Duox2 participates in TLR4-mediated dysbiosis, and will use mucosa-associated microbial transplants to test the tumorigenic effects of this microbiota in germ-free mice. 3) Dissect the role of TLR4 in shaping the response of MDSCs to the resident microbiome during tumorigenesis. We will use dual RNA sequencing of bacterial and host RNA from tumor-associated MDSCs and MDSCs from UC patients to determine the interplay between the resident microbiota and tumor-promoting gene expression in MDSCs. We will also functionally characterize the pro- tumorigenic responses of MDSCs upon TLR4 stimulation including cytokine, chemokine, and H2O2 production. The work proposed herein will provide the mechanistic justification for subsequent human studies to target TLR4 or downstream signaling molecules as a means to halt progression from UC/inflammation to dysplasia and dysplasia to cancer.